modulation of survival signaling pathways and persistence of the

Modulation of Survival Signaling Pathways and Persistence of the

Genotoxic Stress as a Basis for the Synergistic Interaction

between the Atypical Retinoid ST1926 and the Epidermal

Growth Factor Receptor Inhibitor ZD1839

Chiara Zanchi, Valentina Zuco, Cinzia Lanzi, Rosanna Supino, and Franco ZuninoIstituto Nazionale per lo Studio e la Cura dei Tumori, Milan, Italy

Abstract

Strategies targeting apoptotic pathways may have relevance toimprove the efficacy of antitumor therapy. Because syntheticatypical retinoids are potent inducers of apoptosis, there is anincreasing interest in exploiting their potential in noveltherapeutic approaches. In the present study, we haveinvestigated the cellular effects of the combination of a novelatypical retinoid, ST1926, and the epidermal growth factorreceptor inhibitor ZD1839. The results indicated a synergisticinteraction between the two drugs associated with a dramaticenhancement of apoptotic response, up-regulation of thecell death receptor DR5, and caspase 8 activation. Othermolecular events induced by the cotreatment included (a) astabilization of the ST1926-induced genotoxic stress detectedby formation of phosphorylated ;-H2AX foci and (b) a com-plete inhibition of extracellular signal-regulated kinase 1//2(ERK1//2) phosphorylation associated with activation of theproapoptotic protein BAD (i.e., inhibition of phosphorylationon Ser112). In addition, ZD1839 itself inhibited survivalpathways by causing a partial dephosphorylation of Akt anda marked down-regulation of survivin. The role of ERK-mediated survival pathways in the cellular response to thedrug combination was further supported by the counteractingeffect of stimulation of survival pathways by an alternativereceptor tyrosine kinase and by the use of a specific inhibitor ofthe ERK pathway. In conclusion, the results support that thesurvival pathways activated by epidermal growth factorreceptor are determinants of the cell susceptibility toST1926-induced apoptosis and lowering survival signals mayincrease the cellular sensitivity to the atypical retinoid. Thefavorable pharmacologic profiles of both ST1926 and ZD1839suggest that the combination of these well-tolerated agentsmay have therapeutic potential. (Cancer Res 2005; 65(6): 2364-72)

Introduction

Apoptosis is recognized to be an important phenomenon inantitumor drug-induced tumor cell killing and susceptibility toapoptosis of tumor cells is a determinant of efficacy ofchemotherapy (1). Apoptosis regulating programs may beimpaired in human tumors and the fate of treated cells is likelydetermined by the balance between proapoptotic and antiapop-totic signals. Thus, modulation of apoptosis-related pathways

may be a promising approach to improve the efficacy ofantitumor therapy.Synthetic retinoids, which are reported as potent inducers of

apoptosis, represent an emerging class of potentially useful agents(2–5). Whereas the biological effects of natural retinoids aremediated by the retinoid receptors, the proapoptotic activity ofatypical synthetic retinoids seems to be independent of retinoidreceptors (4, 6). We have recently reported that a novel adamantylretinoid, ST1926, is a potent inducer of apoptosis in a variety ofhuman tumor cells (7). As for other retinoid-related molecules,the molecular mechanisms involved in apoptosis induction arenot clearly defined. In ovarian carcinoma cells several features ofthe cellular response to ST1926 are reminiscent of the genotoxicstress response because it involves activation of p53 andmodulation of p53-target genes and genes involved in DNAdamage response (8). Moreover, exposure to ST1926 results inactivation of mitogen-activated protein (MAP) kinases, inparticular the c-jun-NH2-kinase and p38 stress kinase. Theactivation of MAP kinases may have a complex and controversialinfluence in determining the ultimate fate of the cell, dependingon the cell type and the context of various signals received by thecell. In ovarian carcinoma cells, MAP kinases, includingextracellular signal-regulated kinases (ERK), seem to have aprotective role against ST1926-induced cytotoxicity (8). Indeed,these kinases are involved in regulating a number of cellularprocesses including survival/proliferation and could potentiallycontribute to influence the ability of cell to survive in stressconditions. Akt is another protein kinase implicated in thecellular response to various stimuli capable of inducing apoptosis(9) and plays a key role in promoting cell survival throughintegration of pathways triggered by a number of signalsincluding those mediated by growth factor receptors.Epidermal growth factor receptor (EGFR) is a receptor tyrosine

kinase frequently deregulated in cancer cells and has emerged as amolecular target for anticancer therapy (10). EGFR targeting agentsinclude monoclonal antibodies against the extracellular ligand-binding domain of the receptor and small-molecule inhibitors thatblock the activation of the EGFR tyrosine kinase domain. In thesignaling cascade activated by EGFR, the ERKs and Akt play arelevant role regulating cellular processes such as proliferation andsurvival. Thus, EGFR represents a potential target for modulationof downstream protective events. Indeed, tyrosine kinase inhibitorsof the EGFR, such as ZD1839 (Iressa, Gefitinib), have been reportedto interact synergistically with a number of cytotoxic agentsincluding DNA damaging agents (11–14).Therefore, on the basis of evidence that ST1926 induces a DNA

damage response and ST1926-induced cytotoxicity can be modu-

lated by inhibitors of survival pathways (7, 8), the present study wasdone to investigate the cellular effects of a combination of ST1926

Requests for reprints: Franco Zunino, Istituto Nazionale Tumori, Via Venezian 1,20133 Milan, Italy. Phone: 39-02-23902267; Fax: 39-02-23902692; E-mail:[email protected].

I2005 American Association for Cancer Research.

Cancer Res 2005; 65: (6). March 15, 2005 2364 www.aacrjournals.org

Research Article

Research. on April 12, 2019. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

and ZD1839. The results provide evidence of a synergisticinteraction between the two drugs. Inhibition of survival pathways

was accompanied by a persistence of genotoxic damage and an

enhancement of the apoptotic response.

Materials and Methods

Cell Culture and Drugs. The human ovarian carcinoma cell lines

(IGROV-1 and SKOV-3), the vulvar squamous carcinoma cell line A431, thebreast carcinoma cell line MCF-7, and the lung adenocarcinoma cell line

A549 were grown in RPMI 1640 (Bio Whittaker, Verviers, Belgium)

containing 10% fetal bovine serum (Life Technologies, Inc., Gaithersburg,MD) in 5% CO2 at 37jC.

The chemical structure and synthesis of the atypical retinoid ST1926

were previously described (7). The compound (Sigma-Tau, Pomezia, Italy)

was dissolved in DMSO before further dilution in culture medium. Stocksolutions were stored at �20jC. ZD1839 (Iressa) was kindly provided by

Astra Zeneca (Macclesfield, Cheshire, United Kingdom). A 10 mmol/L

solution in DMSO was prepared before use. Final concentration of DMSO in

calcium medium was 0.5%. In all experiments involving hepatocyte growthfactor (HGF; Sigma, St. Louis, MO), cells were exposed to the growth factor

(20 ng/mL) 2 hours before treatment with ZD1839 and ST1926. The MEK1/2

inhibitor UO126 (Sigma) was dissolved in DMSO and added to the culturemedium 30 minutes before treatment with ST1926.

Antiproliferative Activity. Cells were seeded in duplicate into six-well

plates. After 24 hours, cells were exposed to ZD1839, ST1926, or to bothdrugs simultaneously for 72 hours. After treatment, adherent cells were

trypsinized and counted by a cell counter (Coulter Electronics, Luton,

United Kingdom). Drug concentrations producing 10%, 30%, 50%, and 80%

of cell growth inhibition (IC10, IC30, IC50, and IC80, respectively) wereassessed from dose-response curves. Analysis of drug interaction was done

by a modified method of Drewinko et al. (15). Drewinko index >1 indicates

greater than additive effects (synergism; the higher the value, the greater the

degree of synergy), Drewinko index equal to 1 indicates additivity, andDrewinko index less than 1 indicates antagonism. When indicated, the

method described by Chou and Talalay (16) was also applied. In all

experiments, control cells were treated with the solvent at the sameconcentration used in drug-exposed cells.

Cell Cycle and Apoptosis. For cell cycle analysis, cells were trypsinized,fixed in 70% ethanol, stained in phosphate buffered solution (PBS)containing 10 Ag/mL propidium iodide (Sigma) and RNase A (66 units/

mL; Sigma) for 18 hours, and analyzed by FACScan flow cytometer (Becton

Dickinson, Mountain View, CA).

For apoptosis detection, adherent and floating cells were harvested andanalyzed for DNA fragmentation by terminal deoxynucleotidyl transferase–

mediated nick end labeling (TUNEL) assay (Roche, Mannheim, Germany) as

previously described (8). Apoptosis was assessed by flow cytometry and the

results were analyzed using the CellQuest software (Becton Dickinson).

Western Blot Analysis. Adherent and floating cells were lysed in hot

SDS sample buffer [0.125 mol/L Tris-HCl (pH 6.8), 5% SDS, 1 mmol/L

phenylmethylsulfonyl fluoride, 10 Ag/mL pepstatin, 12.5 Ag/mL leupeptin,100 Komberg international units of aprotinin, 1 mmol/L sodium

orthovanadate, 1 mmol/L sodium molybdate] and whole cell lysates were

prepared for SDS-PAGE as described (8). After separation on SDS-PAGE,proteins were transferred to nitrocellulose filters. Membranes were

incubated with primary antibodies overnight and with peroxidase-

conjugated anti-mouse or anti-rabbit antibodies to reveal immunoreactive

bands using the enhanced chemiluminescence detection system fromAmersham Bioscences (Amersham, United Kingdom) or from Pierce

(Rockford, IL).

The following primary antibodies were used: anti-phospho-p44/42

MAP kinase (Thr202/Tyr204; New England Biolabs, Beverly, MA); anti-Bcl-2,anti-caspase 9, anti Met (C-12; Santa Cruz Biotechnology, Santa Cruz,

CA); anti-MAP kinase 1/2 (ERK1/2-CT), anti-EGFR, anti-phospho-Met

(Y1234, Y1235; Upstate Biotechnology, Lake Placid, NY); anti-phospho-Akt

(Ser473), anti-cleaved poly(ADP-ribose) polymerase (PARP), anti-cleaved

caspase 3, anti-phospho-Bad (Ser112; Cell Signaling Technology, Beverly, MA);anti-PBKa/Akt, anti-Bad (Transduction Laboratories, Lexington, KY); anti-

caspase 8, anti-Bcl-X (PharMingen, Becton Dickinson); anti-phospho-EGFR

pY1086 (Biosource, Camarillo, CA); anti-survivin (ab469; Abcam, Cambridge,

United Kingdom); anti-actin and anti-tubulin (Sigma).

Analysis of DR4 and DR5 Expression. For the analysis of death

receptors, cells were harvested and incubated at 4jC with biotinylated-

conjugated antihuman tumor necrosis factor–related apoptosis-inducingligand receptor 1 (TRAIL R1; DR4) or antihuman TRAIL R2 (DR5)

monoclonal antibody (R&D Systems, Minneapolis, MN) for 30 minutes,

and then for 1 hour with streptavidin-phycoerythrin conjugate

(BD PharMingen, Heidelberg, Germany). After washing, cells wereresuspended in PBS and antibody binding was detected by flow cytometry

(FACScan). Cells incubated only with the phycoerythrin conjugate served as

negative controls.

Assessment of Genotoxic Damage. Phosphorylated H2AX histone

(g-H2AX) at the DNA damage sites was detected by immunofluorescence

staining (17). Cells were grown on coverslips, treated with drugs asindicated, and fixed in 2% paraformaldehyde in PBS for 5 minutes. After

washing in PBS, cells were permeabilized in methanol at �20jC for

20 minutes and blocked with PBS containing 1% bovine serum albumin

and 0.1% Tween 20 (PBA) for 15 minutes. Samples were then incubatedwith anti g-H2AX mouse monoclonal antibody (Upstate Biotechnology) for

1 hour, followed by AlexaFluor 594-conjugated goat anti-mouse antibody

(Molecular Probes, Eugene, OR) for 1 hour. Slides were then washed in

PBA, incubated with 2 mg/mL Hoechst 33341 (Sigma) for 2 minutes,mounted with Mowiol, and examined by a fluorescence microscope.

Cells displaying at least 10 foci were considered as positive.

Results

Cell Growth Inhibition. The combination of the EGFR inhibitorZD1839 and the atypical retinoid ST1926 was investigated in apanel of human cancer cell lines of different tissue origin andcharacterized by a differential sensitivity to ZD1839. Using asimultaneous combination treatment with a moderately antipro-liferative concentration of ZD1839 (IC30) and two concentrations ofST1926, a supra-additive effect was consistently observed in alltested cell lines (Table 1). Owing to the synergistic interaction beingmore marked in the IGROV-1 ovarian carcinoma cell line, this cellline was chosen for further studies.

Table 1. Analysis of combined treatment of ST1926 andZD1839 (72 hours) in a panel of human cancer cell lines

Cell lines ZD1839 (IC30) ST1926 D.I.*

IGROV-1 0.04 0.15c

1.800.04 0.08

b1.95

SKOV-3 3 0.14c

1.16

3 0.1b

1.31A431 0.01 0.094

c1.50

0.01 0.08b

1.5

MCF-7 5 0.1c

1.80

5 0.05b

1.71A549 2.5 0.15

c1.60

2.5 0.08b

1.15

*D.I., Drewinko index calculated for the drug combination. D.I. > 1

indicates synergistic interaction.cST1926 at the IC50 as single agent.bST1926 at the IC30 as single agent.

Interaction between ST1926 and ZD1839

www.aacrjournals.org 2365 Cancer Res 2005; 65: (6). March 15, 2005



The dose-response curves for the antiproliferative effects ofST1926 and ZD1839 on IGROV-1 cells treated for 72 hours areshown in Fig. 1A . Cells were treated simultaneously with ST1926 (inthe concentration range of 0.008-0.3 Amol/L) and ZD1839 at toxicand sub-toxic concentrations (i.e., 0.01, 0.04, and 0.1 Amol/Lcorresponding to the IC10, IC30, and IC50, respectively). A supra-additive growth inhibitory effect was clearly documented by thedose-response curve of ST1926 in the presence of a sub-toxic doseof ZD1839 (0.01 Amol/L). The analysis of drug interaction (15)confirmed that the ST1926/ZD1839 combination was synergic(Fig. 1B). Drug interaction investigated and analyzed according toChou and Talalay (16) supported the synergism (not shown).Cell Cycle Analysis and Apoptosis. The analysis of cell cycle

perturbations was done on cells exposed to each agent, ST1926(IC50) or ZD1839 (IC50 and IC30), or to their combination.Simultaneous exposure to the two drugs for 72 hours did notdetermine modifications of the cell cycle distribution comparedwith that of control cells or of cells exposed to each drug (Fig. 2A).However, a marked increase in the sub-G1 DNA content was foundin cells exposed to the drug combination (4-5% in ZD1839- orST1926-treated cells versus 30% and 42% in cells treated with thecombination at IC30 and IC50 of ZD1839, respectively), thussuggesting the induction of apoptotic cell death.To determine whether the drug interaction resulted in an

enhancement of apoptosis, cells were further analyzed by theTUNEL assay (Fig. 2B). A marginal induction of apoptosis wasobserved in the presence of each single agent, whereas thecombined treatment produced a remarkable increase of apoptosiswhich was dependent on the ZD1839 concentration.Modulation of Apoptotic Pathways. Apoptotic pathways

activated by the combined treatment were investigated by Westernblot analysis. Cells were treated with ST1926 or ZD1839, alone or incombination, at concentrations corresponding to the IC50 values(0.15 and 0.1 Amol/L, respectively) for 48 or 72 hours. In theseconditions, barely detectable caspase activation was induced by thetwo drugs given alone (Fig. 3) according to a marginal induction ofapoptosis detected by the TUNEL assay (Fig. 2B). By contrast, incells treated with the drug combination, a marked cleavage ofcaspase 8, caspase 3, and, to a lesser extent, caspase 9 wasassociated with cleavage of the caspase substrate PARP, thus

confirming an enhanced activation of caspase-dependent apoptoticpathways. To get insights into the mechanisms responsible forenhancement of apoptosis by the cotreatment, we examined theexpression of negative regulators of apoptosis belonging toinhibitors of apoptosis protein family and to the Bcl-2 family.ZD1839 produced a marked reduction in survivin levels after 48and 72 hours of treatment (Fig. 3). Such an effect, alreadydetectable after 24 hours of treatment (not shown), wasmaintained in the combined treatment. Similar effects, althoughto a lesser degree, were observed on the expression levels of Bcl-2and Bcl-XL (not shown).Effect on the Expression of Death Receptors. Owing to

caspase 8, which is implicated in cell death receptor-mediatedapoptosis, markedly activated by the combination of ST1926 andZD1839, the drug effects on the expression of death receptors DR4and DR5 were examined. Whereas ST1926 or ZD1839 alone did notenhance the expression of DR5 (Fig. 4A), an up-regulation of thisreceptor was found after simultaneous treatment with ST1926 plusZD1839 for 48 hours (Fig. 4B). The synergistic effect on thisreceptor likely reflected the high level of caspase 8 activation andtherefore the involvement of the extrinsic pathway in the apoptoticresponse. No modulation of DR4 expression was found (notshown).Effects on Proliferation//Survival Pathways. To investigate the

nature of the cooperation between ZD1839 and ST1926, weexamined the effects of the two drugs on proliferation/survivalpathways sustained by the EGFR signaling. At first, to check theeffect of ZD1839 on its target in our experimental conditions, weexamined the tyrosine phosphorylation of EGFR in IGROV-1 cellstreated with the two compounds alone or in combination. Asshown in Fig. 5A , EGFR autophosphorylation was inhibited aftertreatment with ZD1839 (IC50) and this effect was maintained whencells were exposed to ZD1839 and ST1926 simultaneously for24 hours. ST1926 alone did not affect EGFR autophosphorylation.The activation state of ERKs and Akt was examined in cells

exposed to ST1926 (IC50) and/or to ZD1839 (IC50 and IC80), for 6 or24 hours. Akt activation, detected as Ser473 phosphorylation, wasnot modulated by ST1926. ZD1839 appreciably reduced the level ofAkt phosphorylation following 24 hours of treatment (53% and 23%versus control with 0.1 and 1 Amol/L ZD1839, respectively) and the

Figure 1. Effect of ST1926 (ST ) onIGROV-1 cell growth in the absence (n) orpresence of 0.01 Amol/L (E), 0.04 Amol/L(!), and 0.1 Amol/L (5) ZD1839 (ZD ).Antiproliferative effect of ZD1839 alone(o). A, dose-dependent inhibitory effectsafter 72 hours of treatment. B, combinationindex calculated according to theDrewinko test.

Cancer Research




combination with the atypical retinoid did not essentially modifythis effect (Fig. 5B). A marked dose-dependent inhibition of ERK1/2phosphorylation was induced by ZD1839, already evident after6 hours of treatment and more pronounced after 24 hours. Theactivation of the two ERKs was not substantially modulated by

ST1926 alone. Nonetheless, the combination with ZD1839 inducedan apparent supra-additive inhibition of ERKs phosphorylation ateither 6 or 24 hours (Fig. 5B).Role of Survival//Proliferation Pathways. To provide addi-

tional support to the role of survival/proliferation pathways in thesynergistic interaction between ZD1839 and ST1926, we didcombination experiments in the presence of HGF as a survival/proliferation stimulus. Cellular response to HGF stimulation isknown to protect cells from apoptosis, induced by a variety ofstimuli, through activation of the PI3K/Akt and ERK pathways(18). Therefore cellular response to HGF was determined by theinduction of phosphorylation of Met receptor, ERKs, and Akt(Fig. 6A). The stimulation with the growth factor resulted in lossof the synergistic interaction of the drugs at their respective IC50

and strongly reduced apoptosis after 72 hours of combinationtreatment (Fig. 6D). In addition, HGF attenuated the inhibitoryeffects of the drugs on activation of ERKs and, to a lesser extent,of Akt observed after 24 hours of treatment (Fig. 6B). Owing tothe proapoptotic factor BAD, a Bcl-2 family member, being atarget of both the MEK/ERK and the PI3K/Akt pathways, weinvestigated its phosphorylation status. Indeed, phosphorylationof BAD on the specific serine residues, Ser112 (through the ERKsignaling) or Ser136 (through Akt), is known to suppress itsproapoptotic function by inhibiting the interaction with theantiapoptotic proteins, Bcl-2 and Bcl-XL (26). According to theeffect of ZD1839 and ST1926 on ERK and Akt activation, BADphosphorylation on Ser112 was reduced by ZD1839 either alone orin combination (Fig. 6B) and phosphorylation on Ser136 seemedbarely affected (not shown). Again, the effects of the two drugswere attenuated in the presence of HGF. The counteracting actionof HGF on the biochemical effects of the drug combination wasstill evident after 72 hours of treatment (Fig. 6C). In addition,

Figure 2. Cell cycle distribution (A ) and apoptosis (B) in IGROV-1 cells treated with ST1926 (IC50: 0.15 Amol/L) and ZD1839 (IC50: 0.1 Amol/L and IC30: 0.04 Amol/L),alone or in combination, for 72 hours. Cell cycle and apoptosis were assessed by fluorescence-activated cell sorting analysis on propidium iodide- orTUNEL-stained cells, respectively. The percentages of cells in the different cell cycle phases (A ) and the percentage of TUNEL-positive cells (B ) are indicated.Representative of at least three independent experiments.

Figure 3. Cleavage of caspases and PARP and modulation of survivinexpression. Cells were treated with ST1926 or ZD1839, alone or in combination,at their IC50 (0.15 and 0.1 Amol/L, respectively) for 48 and 72 hours. Whole-celllysates were analyzed by Western blotting with the indicated specific antibodies;actin is shown as control for protein loading.





in cells exposed to the growth factor, drug-induced cleavage ofcaspases (8 and 3) and PARP was decreased together with areduced down-regulation of survivin.Owing to the above results indirectly supporting that the

enhancement of apoptosis induced by the ST1926/ZD1839combination was mainly related to the enhanced inhibition ofthe survival pathways involving the ERKs, we explored theeffects of ERK pathway inhibition by an alternative approach.ST1926 treatment was done in the presence of the MEK1/2inhibitor UO126. Figure 7 shows that, in a range of UO126concentrations inhibiting ERK activation, a supra-additive effecton induction of caspases and PARP cleavage was obtained incells exposed to the ST1926/UO126 combined treatment ascompared with each single-agent treatment. Moreover, a moremarked survivin down-regulation was observed in cells exposedto the cotreatment as compared with treatment with the ERKinhibitor alone. Altogether, these results support a major role ofsurvival/proliferation pathways in the modulation of cellresponse to the ST1926/ZD1839 combination, and indicate thatinhibition of the ERK pathway by the EGFR inhibitor provides arelevant contribution to the synergistic interaction between thetwo drugs.

Genotoxic Damage. To analyze whether EGFR blockade finallyaffected the genotoxic damage induced by ST1926, we examinedthe effects of the cotreatment with ZD1839 on the phosphorylationof H2AX histone (g-H2AX), which can be visualized as foci at theDNA damage sites by immunofluorescence with phospho-specificantibodies. Indeed, this approach is the most sensitive method todetect double-strand DNA breaks (19, 20). Cells were treated withST1926 (IC50) or ZD1839 (IC80) alone or in combination andwashed after 3 hours of drug exposure. A short-term exposurewas employed to avoid overlapping of apoptotic DNA fragmen-tation. g-H2AX foci were examined after the drug washout andafter 3 hours of recovery in the presence or absence of ZD1839(Fig. 8A). No g-H2AX foci were observed in control cells (Fig. 8A)and in ZD1839 treated cells (not shown). About 40% of cellswere highly positive for g-H2AX after treatment with ST1926alone or in combination with ZD1839 (Fig. 8B). Such DNAlesions were reversible upon drug removal in cells treated withST1926 alone. In contrast, in the cells exposed to the combinedtreatment, g-H2AX positivity was still evident in 25% to 30% ofcells in the presence of ZD1839 during the recovery time. Theabsence of ZD1839 during the 3 hours recovery resulted in a lowlevel of positivity (3%). These results suggest that ZD1839 is able

Figure 4. Expression of DR5 in IGROV-1cells treated for 48 hours with the IC50

of ST1926 (0.15 Amol/L) and ZD1839(0.1 Amol/L), single (A ) or combined (B).A, profiles obtained from control cells andST1926-treated cells are superimposed.Representative of two independentexperiments.

Figure 5. Inhibition of EGFR autophosphorylation anddownstream pathways detected by Western blot analysis in cellsexposed to ST1926 and ZD1839 in single or combined treatment.A, EGFR autophosphorylation in cells treated with ST1926 andZD1839 at their respective IC50 (0.15 and 0.1 Amol/L) for 24 hoursdetected using a phospho-specific (Y1086) antibody. B,dose-dependent inhibition of activating phosphorylations of Akt(Ser473) and ERKs (Thr202/Tyr204) after cell exposure to the drugs(0.15 Amol/L ST1926; 0.1 and 1 Amol/L ZD1839) for 6 or 24 hours.Filters were stripped and reprobed with antibodies recognizing therespective proteins (EGFR, Akt, and ERKs) or tubulin to controlspecific or overall protein loading.

Cancer Research




to promote a persistence of ST1926-induced genotoxic stress.Analysis of DNA fragmentation detected by alkaline elution techniquealso supported the accumulation of DNA lesions in the presence ofboth agents (data not shown).

Discussion

The ability of tumor cells to survive under stress conditions (e.g.,cytotoxic treatment) may critically influence their sensitivity andresponse to therapy. Resistance to drug-induced apoptosis mayreflect defects in the apoptotic programs and/or over activation ofprotective signal transduction pathways which promote survival.For example, ERK activation has been implicated as a survivalfactor and constitutive activation of MEK/ERK cascade found inseveral tumors is likely involved in supporting malignantprogression and proliferation (21). Again, several components ofthe antiapoptotic PI3K/Akt pathway are deregulated in a widespectrum of human cancers contributing to promote cell survival(22). Because of these protective pathways also involved in the

cellular stress response, in particular in DNA damage response,apoptosis induced by cytotoxic agents could be enhanced byinhibition of growth factor receptor–mediated signal transductionpromoting survival. In this context, EGFR might represent apotential target not only for the control of tumor growth but alsoto improve the efficacy of cytotoxic agents (23, 24).In the present study, we obtained evidence that in ovarian

carcinoma cells treated with ST1926, the concomitant inhibition ofsurvival signaling pathways by ZD1839 resulted in a markedinduction of apoptosis and in synergistic cytotoxicity. AlthoughST1926 is not a conventional cytotoxic agent, several direct andindirect lines of evidence support the drug ability to induce agenotoxic stress (7, 8). Indeed, the pattern of cellular response andmodulation of genes implicated in DNA damage response supportthat ST1926-induced genotoxic stress is implicated in mediatingapoptosis induction (8). Moreover, early detection of DNA breaks,p53 activation, and up-regulation of p53-target genes in treatedcells are consistent with this interpretation. The present studyprovides additional support to the proposed mechanism involving

Figure 6. Cellular and biochemical effects of ST1926/ZD1839 combined treatment in the presence of stimulation with HGF. IGROV-1 cells were exposed to the growthfactor (20 ng/mL) 2 hours before treatment with ST1926 (0.15 Amol/L) or ZD1839 (0.1 Amol/L) or their combination. A, activation of MET, ERKs, and Akt after cellexposure to HGF for 2 hours. B, modulation of ERK, Akt, and BAD activation after 24 hours of treatment. C, modulation of the same proteins as in B , in addition tocaspases, PARP, and survivin, after 72 hours of treatment. A-C, cell lysates were processed for Western blot analysis with the indicated antibodies. D, apoptosisinduction after 72 hours of treatment. The percentages of TUNEL-positive cells are indicated in each panel; D.I., Drewinko index value referred to the antiproliferativeeffects of the combined treatments.





the formation of DNA lesions as a primary event in apoptosisinduction by ST1926. Indeed, we present evidence that ZD1839/ST1926 combination was able to induce an early DNA damagedetectable as g-H2AX foci under conditions producing a synergisticeffect on apoptosis induction. Using this approach, we found thatZD1839 produced a persistence of genotoxic lesions induced byST1926, which were otherwise reversible following drug removal. Aproposed rationale for the observed synergistic effects betweenZD1839 and DNA-damaging agents is that DNA repair processesmight be impaired by the EGFR inhibitor (12, 13). Indeed, bypromoting cellular proliferation, EGFR activation might also favorDNA repair. The molecular determinants of such a connection arenot clearly defined. However, recent evidence suggests theinvolvement of an ERK-dependent pathway in this process (25).ERK activation plays a central role in signal transduction, due to itsimplication in pathways activated by growth factor receptors and,therefore, in promoting survival and proliferation. We previouslyreported that the inhibition of the MEK/ERK pathway by the MEKinhibitor PD98059 enhanced the antiproliferative and proapoptoticeffects induced by ST1926 (8). Here, we report that ZD1839, apotent inhibitor of EGFR tyrosine kinase activity, reduced in a dose-dependent manner the phosphorylation of the ERKs, and thisinhibition was strengthened in combination with ST1926. Thiseffect resulted in activation of BAD, as a result of dephosphory-lation on Ser112, which is known to be phosphorylated downstreamto the ERKs (26). BAD activation could be further enhanced bypartial dephosphorylation on Ser136 (not shown) as a result of apartial inhibition of Akt pathway by ZD1839. The experiments donewith the MEK1/2 inhibitor UO126 again support the involvement ofinhibition of ERK-mediated pathways in sustaining the enhance-ment of apoptosis in the drug combination. Consistent with suchinterpretation is the observation that in conditions in which theactivation of survival/proliferation pathways, including ERKs andAkt, was promoted by a receptor alternative to EGFR (i.e., the Metreceptor), apoptosis induced by the ZD1839/ST1926 combinationwas strongly reduced and the synergistic interaction was lost.Although Akt pathway is known to play a role in cell survival and

its activation is mediated through growth factor receptors, theeffect of ZD1839 on this pathway was less marked. A recent studyhas reported that a closely related compound, 3-Cl-AHPC, was ableto decrease Akt kinase activity associated with EGFR proteolysis inprostate and breast carcinoma cells (27). However, in ourexperimental conditions we did not detect any effect of ST1926itself or ST1926/ZD1839 combination on the level of EGFR protein.In addition to BAD, other apoptosis-related proteins are known

to be regulated by the ERK MAP kinases and inhibition of MEK/ERK pathway has downstream effects on antiapoptotic proteins(28–30). Our results indicate that inhibition of EGFR-ERKpathway by ZD1839 or UO126 produced a down-regulation ofBcl-2 and Bcl-XL (not shown), as previously reported (13, 14), andof survivin, which were maintained in the combined treatment.Although ST1926 is known to activate p53 (8) and p53 mayrepress the transcription of survivin (31), the reduced expressionof the survivin protein was an effect of ZD1839 alone. Survivindown-regulation, as a consequence of inhibition of the MEK/ERKsignaling, was previously described (28). Our finding is the firstevidence that survivin down-regulation can be induced by anEGFR inhibitor. Survivin is involved in the regulation of both celldeath and proliferation because it not only protects cells fromapoptosis as an endogenous repressor of the mitochondrialpathway but it is also implicated in cell division and plays a rolein the mitotic checkpoint following cellular damage (32).

Therefore, survivin overexpression in tumor cells may representa mechanism of resistance to genotoxic agents (33). Our data thussuggest that the synergistic interaction between ZD1839 andST1926 implicates an enhancement of the intrinsic mitochondrial-regulated pathway with BAD activation and down-regulation ofsurvivin and Bcl-2/Bcl-XL. In addition, a relevant cooperation bythe extrinsic pathway was indicated by the DR5 death receptorup-regulation associated with caspase 8 activation. Owing to DR5as a DNA-damage inducible gene (34), it is conceivable that asustained activation of this pathway might be favored by thepersistent DNA damage in cells exposed to the cotreatment. Theactivation of apoptotic signaling pathway mediated by deathreceptors is recognized to play an important role in atypicalretinoid-induced apoptosis (35).In conclusion, the synergistic antiproliferative and apoptotic

effects exhibited by the combination of the atypical retinoidST1926 and the EGFR receptor inhibitor ZD1839 were related toinhibition of protective survival pathways, in particular ERK-mediated processes, and likely to modulation of cellular pathwayscontrolling DNA repair. The activation of survival signalingpathways mediated by EGFR seems to be a critical event indetermining the susceptibility to ST1926-induced apoptosis.Finally, our results support that targeting survival pathways mayhave pharmacologic implications in an attempt to improve theefficacy of stress-inducing agents. Although clinical studies ofZD1839/platinum compound combinations have been negativeand the clinical use of ZD1839 has shown efficacy only as asingle-agent therapy in a subset of treated patients (36), recentprogress in understanding the molecular basis of clinical

Figure 7. A, dose-dependent inhibition of ERKs phosphorylation in IGROV-1cells exposed to the MEK1/2 inhibitor UO126 for 72 hours. B, cleavage ofcaspases and PARP and modulation of survivin expression after cell exposure toST1926 (0.15 Amol/L) or UO126 (2, 3, and 5 Amol/L), or their combination, for72 hours. Western blot analysis was done with the indicated antibodies. ERKsand actin blots are shown as control for protein loading.

Cancer Research




responsiveness to this drug suggests the possibility of selectingpatients who likely could benefit from treatment with the drugalone and, possibly, in rationally designed combinations (37, 38).In such a context, the combination of ZD1839 and ST1926 mayhave a therapeutic potential considering the good tolerabilityprofiles of both agents.

Acknowledgments

Received 7/30/2004; revised 12/21/2004; accepted 1/4/2005.Grant support: Associazione Italiana Ricerca sul Cancro, Milan, Ministero della

Salute, Rome, and Consiglio Nazionale delle Ricerche, Rome, Italy.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.

Figure 8. Effects of ZD1839 on DNAdamage induced by ST1926 in IGROV-1cells. Cells were treated for 3 hours withST1926 alone or with ZD1839/ST1926combination. After drug washout,ST1926-treated cells were incubated foradditional 3 hours in the absence of thedrug. Cells treated with the combinationwere reincubated for 3 hours in thepresence (+ZD ) or absence (�ZD ) ofZD1839. A, immunofluorescence stainingof g-H2AX foci (red). Nuclei werestained with Hoechst 33342 (blue ). B,quantification of cells exhibiting g-H2AXfoci. The percentages were obtained byfluorescence microscope counting fromsmears prepared as in A . Cells displaying10 or more foci were counted as positive.

References1. Hu W, Kavanagh JJ. Anticancer therapy targeting theapoptotic pathway. Lancet 2003;4:721–9.

2. Garattini E, Gianni M, Terao M. Retinoid relatedmolecules an emerging class of apoptotic agents withpromising therapeutic potential in oncology: pharma-cological activity and mechanisms of action. CurrPharm Des 2004;10:433–48.

3. Fontana JA, Rishi AK. Classical and novel retinoids:their targets in cancer therapy. Leukemia 2002;16:463–72.

4. Ortiz MA, Bayon Y, Lopez-Hernandez FJ, Piedrafita FJ.Retinoids in combination therapies for the treatment of

cancer: mechanisms and perspectives. Drug ResistUpdat 2002;5:162–75.

5. Holmes WF, Soprano DR, Soprano KJ. Syntheticretinoids as inducers of apoptosis in ovarian carcinomacell lines. J Cell Physiol 2004;199:317–29.

6. Lotan R. Receptor-independent induction of apoptosisby synthetic retinoids. J Biol Regul Homeost Agents2003;17:13–28.

7. Cincinelli R, Dallavalle S, Merlini L, et al. A novelatypical retinoid endowed with proapoptotic andantitumor activity. J Med Chem 2003;46:909–12.

8. Zuco V, Zanchi C, Cassinelli G, et al. Induction ofapoptosis and stress response in ovarian carcinoma cell

lines treated with ST1926, an atypical retinoid. CellDeath Differ 2004;11:280–9.

9. Martindale JL, Holbrook NJ. Cellular response tooxidative stress: signaling for suicide and survival.J Cell Physiol 2002;192:1–15.

10. Mendelsohn J, Baselga J. The EGF receptor family astargets for cancer therapy. Oncogene 2000;19:6550–65.

11. Koizumi F, Kanzawa F, Ueda Y, et al. Synergisticinteraction between the EGFR tyrosine kinase inhibitorgefitinib (‘‘Iressa’’) and the DNA topoisomeraseI inhibitor CPT-11 (irinotecan) in human colorectalcancer cells. Int J Cancer 2004;108:464–72.

12. Xu J-M, Azzariti A, Montemuro S, Lu B, Colucci G,





Paradiso A. Characterization of sequence-dependentsynergy between ZD1839 (‘‘Iressa’’) and oxaliplatin.Biochem Pharmacol 2003;66:551–63.

13. Magne N, Fischel J-L, Tisson C, et al. Molecularmechanisms underlying the interaction betweenZD1839 (‘‘Iressa’’) and cisplatin/5-fluorouracil. BrJ Cancer 2003;89:585–92.

14. Bianco C, Tortora G, Bianco R, et al. Enhancement ofantitumor activity of ionizing radiation by combinedtreatment with the selective epidermal growth factorreceptor-tyrosine kinase inhibitor ZD1839 (Iressa). ClinCancer Res 2002;8:3250–8.

15. Drewinko B, Loo TL, Brown JA, Gottlieb JA, FreireichEJ. Combination chemotherapy in vitro with Adriamy-cin. Observations of additive, antagonistic, and syner-gistic effects when used in two-drug combination oncultured human lymphoma cells. Cancer BiochemBiophys 1976;1:187–95.

16. Chou TC, Talalay P. Quantitative analysis of dose-effect relationships: the combined effects of multipledrugs or enzyme inhibitors. Adv Enzyme Regul1984;22:27–55.

17. Furuta T, Takemura H, Liao Z-Y, et al. Phosphoryla-tion of histone H2AX and activation of Mre11, Rad50,and Nbs1 in response to replication-dependent DNAdouble-strand breaks induced by mammalian DNAtopoisomerase I cleavage complexes. J Biol Chem2003;278:20303–12.

18. Xiao G-H, Jeffers M, Bellacosa A, Mitsuuchi Y,Vande Woude GF, Testa JR. Anti-apoptotic signaling byhepatocyte growth factor/Met via the phosphati-dylinositol 3-kinase/Akt and mitogen-activated proteinkinase pathways. PNAS 2001;98:247–52.

19. Modesti M, Kanaar R. DNA repair: spot(light)s onchromatin. Curr Biol 2001;11:229–32.

20. d’Adda di Fagagna F, Reaper PM, Clay-Farrace L, et al.A DNA damage checkpoint response in telomere-initiated senescence. Nature 2003;426:194–8.

21. Hoshino R, Chatani Y, Yamori T, et al. Constitutiveactivation of the 41-/43-kDa mitogen-activated proteinkinase signaling pathway in human tumors. Oncogene1999;18:813–22.

22. Nicholson KM, Anderson NG. The protein kinase B/Akt signaling pathway in human malignancy. CellSignal 2002;14:381–95.

23. Mendelson J, Baselga J. Status of epidermal growthfactor receptor antagonists in the biology and treat-ment of cancer. J Clin Oncol 2003;21:2787–99.

24. Levitzki A. EGF receptor as a therapeutic target.Lung Cancer 2003;41:9–14.

25. Yacoub A, McKinstry R, Hinman D, Chung T,Dent P, Hagan MP. Epidermal growth factor andionizing radiation up-regulate the DNA repair genesXRCC1 and ERCC1 in DU145 and LNCaP prostatecarcinoma through MAPK signaling. Radiat Res2003;159:439–52.

26. Gilmore AP, Valentijn AJ, Wang P, et al. Activationof BAD by therapeutic inhibition of epidermalgrowth factor receptor and transactivation by insu-lin-like growth factor receptor. J Biol Chem 2002;277:27643–50.

27. Farhana L, Dawson MI, Huang Y, et al. Apoptosissignaling by the novel compound 3-Cl-AHPC involvesincreased EGFR proteolysis and accompanying de-creased phosphatidylinositol 3-kinase and AKT kinaseactivities. Oncogene 2004;23:1874–84.

28. Wiesenauer CA, Yip-Schneider MT, Wang Y, SchmidtCM. Multiple anticancer effects of blocking MEK-ERKsignaling in hepatocellular carcinoma. J Am Coll Surg2004;198:410–21.

29. Hu Y, Bally M, Dragowska WH, Mayer L. Inhibition ofmitogen-activated protein kinase/extracellular signal-regulated kinase kinase enhances chemotherapeuticeffects on H460 human non–small cell lung cancer cellsthrough activation of apoptosis. Mol Cancer Ther2003;2:641–9.

30. Boucher MJ, Morisset J, Vachon PH, Reed JC, Laine J,Rivard N. MEK/ERK signaling pathway regulates theexpression of Bcl-2, Bcl-X(L), and Mcl-1 and promotessurvival of human pancreatic cancer cells. J CellBiochem 2000;79:355–69.

31. Hoffman WH, Biade S, Zilfou JT, Chen J, MurphyM. Transcriptional repression of the anti-apoptoticsurvivin gene by wild type p53. J Biol Chem 2002;277:3247–57.

32. Altieri DC. Validating survivin as a cancer therapeu-tic target. Nature 2003;3:46–54.

33. Pennati M, Binda M, Colella G, et al. Ribozyme-mediated inhibition of survivin expression increasesspontaneous and drug-induced apoptosis and decreasesthe tumorigenic potential of human prostate cancercells. Oncogene 2004;23:386–94.

34. Wu GS, Kim K, el-Deiry WS. KILLER/DR5, anovel DNA-damage inducible death receptor gene,links the p53-tumor suppressor to caspase activationand apoptotic death. Adv Exp Med Biol 2000;465:143–51.

35. Sun S-Y, Yue P, Hong WK, Lotan R. Augmentation oftumor necrosis factor-related apoptosis-inducing ligand(TRAIL)-induced apoptosis by the synthetic retinoid6-[3-(1-adamantyl)-4-hydroxyphenyl]-2-naphthalenecarboxylic acid (CD437) through up-regulation of TRAILreceptors in human lung cancer cells. Cancer Res 2000;60:7149–55.

36. Grunwald V, Hidalgo M. Developing inhibitors of theepidermal growth factor receptor for cancer treatment.J Natl Cancer Inst 2003;95:851–67.

37. Lynch TJ, Bell DW, Sordella R, et al. Activatingmutations in the epidermal growth factor receptorunderlying responsiveness of non–small-cell lung can-cer to gefitinib. N Engl J Med 2004;350:2129–39.

38. Paez JG, Janne PA, Lee JC, et al. EGFR mutations inlung cancer: correlation with clinical response togefitinib therapy. Science 2004;304:1497–500.

Cancer Research




2005;65:2364-2372. Cancer Res Chiara Zanchi, Valentina Zuco, Cinzia Lanzi, et al. Epidermal Growth Factor Receptor Inhibitor ZD1839Interaction between the Atypical Retinoid ST1926 and theof the Genotoxic Stress as a Basis for the Synergistic Modulation of Survival Signaling Pathways and Persistence

Updated version

http://cancerres.aacrjournals.org/content/65/6/2364

Access the most recent version of this article at:

Cited articles

http://cancerres.aacrjournals.org/content/65/6/2364.full#ref-list-1

This article cites 37 articles, 9 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/65/6/2364.full#related-urls

This article has been cited by 3 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. (CCC)Click on "Request Permissions" which will take you to the Copyright Clearance Center's

.http://cancerres.aacrjournals.org/content/65/6/2364To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/content/65/6/2364.full#ref-list-1

http://cancerres.aacrjournals.org/content/65/6/2364.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



modulation of survival signaling pathways and persistence of the

Documents